Shuman Ghosemajumder dedicated two blog posts to blasting ClickForensics and its latest assessment of click fraud taking place in the pay per click industry.
Who to believe? Shuman has been spoiling for this fight ever since SES San Jose 2006, when he offered some comments about how third parties measure click fraud, first saying in an AdWords blog post the methodology used by ClickForensics and others was flawed, before saying in a conference session that “we’re not saying their methodology is flawed.”
Flawed or not flawed? Shuman says the former in blog post number one about the ClickForensics report, which we mentioned here. If he’s left the door open to try and find some common ground with Tom Cuthbert of ClickForensics, it’s a very slim opening judging by his observations:
We found serious flaws in their counting of clicks – a more fundamental issue than their counting of click fraud. They were making basic counting mistakes and inflating the number of clicks by an average of 40%.
This over-counting problem results in an even more dramatic inflation of click fraud estimates, in fact consistently classifying an advertiser’s best users (the ones spending time browsing their site) as fraudulent.
The gist of Shuman’s posts (part two goes into the technical side) is that Google does so little wrong at filtering out illicit clicks that using third party assessments, which are filled with “egregious mistakes,” actually hurt advertising campaigns based on that data.
Marketing Pilgrim’s Andy Beal talked with Shuman during SES Chicago 2006 about click fraud. Without getting into specific details of determining illicit clicks (because if there were an open discussion about that the bad guys would win), Shuman illustrated how click fraud actually accounted for less than two percent of clicks.
Andy also commented on Shuman’s most recent duo of posts, and cited the need for an independent audit to learn the truth:
Certainly, Google has a lot at stake, so you can argue that they’re just as biased in their defense, as Click Forensics is in their attack. Until we’re able to get independent tracking of click fraud, advertisers will just have to rely on measuring the ROI from a click, and deciding if it makes sense to continue bidding.
The truth, as in most places where there are two sides, probably resides somewhere in the middle. Not as rosy as Shuman’s 2 percent world, nor as gloomy as ClickForensics’ 14.2 percent Mordor.
Shuman did note in response to a commentor on his blog that Google is willing to participate in an audit. They and other PPC providers are working with the IAB Click Measurement Working Group with a goal of setting up just such an approach.
This work has been going on since August 2006, with the first goal being the act of defining a click. Big brand advertisers want audited numbers from interactive publishers by the middle of this year, and measurement certified numbers in 2008.
IAB said in August 2006 that “audited numbers ensure that ad impressions are being counted accurately; while certified numbers indicate that the process supporting the ad counting is consistent and reliable.”
It’s been six months; it might be time for everyone involved to quit foot-dragging on defining a click and get this done so advertisers can finally get some audited data and put the click fraud questions to rest.
UPDATE!: Shuman weighed in with an important clarification to this article; please click here to read it.
—
Tag:
Add to Del.icio.us | Digg | Reddit | Furl
David Utter is a staff writer for murdok covering technology and business.